Fluid Phase Equilibria 337 (2013) 26–31 Contents lists available at SciVerse ScienceDirect Fluid Phase Equilibria journal homepage: www.elsevier.com/locate/fluid A study on thermodynamics effect of [EMIM]-Cl and [OH-C2 MIM]-Cl on methane hydrate equilibrium line Behzad Partoon, Nordiyana M.S. Wong, Khalik M. Sabil ∗ , Khashayar Nasrifar, Mohd Riduan Ahmad Chemical Engineering Department, Universiti Teknologi PETRONAS, 31750 Tronoh, Malaysia a r t i c l e i n f o Article history: Received 29 June 2012 Received in revised form 13 September 2012 Accepted 18 September 2012 Available online 26 September 2012 Keywords: Phase behavior modeling Methane hydrate equilibrium Ionic liquid Experimental measurement a b s t r a c t In this study, the equilibrium conditions of methane hydrate is measured experimentally in the presence of 1-ethyl-3-methyl-imidazolium chloride ([EMIM]-Cl) and 1-hydroxylethyl-3-methyl-imidazolium chloride ([OH-C2 MIM]-Cl) solutions. These two ionic liquids are chosen to study their performances as low dosage hydrate inhibitors. To study the effect of these ionic liquids on the equilibrium phase boundary of methane hydrate, several experiments are conducted in a pressure range of 4–12 MPa. In addition, the equilibrium data in [EMIM]-Cl solutions are modeled using an equation that takes into account the effects of electrolyte on the activity of water. Results show that phase boundary of methane hydrate is shifted toward lower temperature at constant pressure from 0.1 to 1.5 K in the presence of these ionic liquids. This temperature shift, however, becomes more significant at pressures higher than 70 MPa. © 2012 Elsevier B.V. All rights reserved. 1. Introduction Gas hydrates are ice-like crystalline compounds. Gas hydrates are formed by stabilizing water molecules in lattice network using suitable gases at temperatures normally higher than ice point [1]. When Hammerschmidt published his investigation results on gas pipeline plugs, which claimed that the main cause of these plugs is formation of gas hydrate, the issues related to gas hydrates have become more important for oil and gas companies [2]. Since then, efforts for finding gas hydrate inhibitors have seriously started. Nowadays, different types of gas hydrate inhibitors are available in market. These inhibitors are divided into two main groups: thermodynamic hydrate inhibitors (THIs) and low dosage hydrate inhibitors (LDHIs). THIs inhibits hydrate formation in the same way as anti-freezing agents hinder ice formation, that is, hydrate would form at lower temperature at the same pressure. Methanol and glycols are two common THIs that have widely been used in oil and gas industries. THIs are used normally around 20–40 wt%, however, some times higher dosage is recommended [3]. LDHIs, on the other hand, are normally used in smaller quantities, i.e. at ppm level. The LDHI is known to consist of two main categories: kinetics hydrate inhibitors (KHIs) and anti-agglomerate (AA) inhibitors. Recently, Xiao and Adidharma [4] introduced ionic liquids as another class of gas hydrate inhibitors, which are called “dual ∗ Corresponding author. Tel.: +60 5 368 7684; fax: +60 5 365 5670. E-mail addresses: khalik msabil@petronas.com.my, halik98@yahoo.com (K.M. Sabil). 0378-3812/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.fluid.2012.09.025 function inhibitors.” Ionic liquids are organic salts that at room or moderate temperatures are in liquid phase. Xiao and Adidharma [4] found that some ionic liquids could show thermodynamic inhibition and at the same time delay hydrate formation by slowing down the hydrate nucleation rate. This is attributed to their strong electrostatic charges and hydrogen bond with water. This means that ionic liquids are both thermodynamic and kinetic inhibitor. The ionic liquids they initially used are imidazolium cation-based ones. Later, Del Villano and Kelland [5] performed some experiments by two of the previously studied ionic liquids at typical subsea temperatures and subcooling. They reported that these ionic liquids are very weak KHIs at 5000–10,000 ppm concentrations. Xiao et al. [6] continued their investigation with six other dialkylimidazolium halide ionic liquids. These ionic liquids were studied with concentrations about 10 wt% for thermodynamic effects and 1 wt% for kinetics studies. Their results showed a temperature decrease of 0.2–1.2 K in the dissociation temperature of methane hydrate. Among all the ionic liquids studied, 1-ethyl-3methyl-imidazolium chloride ([EMIM]-Cl) was shown to be the most effective thermodynamic inhibitor, while 1-butyl-3-methylimidazolium iodide ([BMIM]-I) was the best kinetic inhibitor. Kim et al. [7] hypothesized that suitable ionic liquid to be used as hydrate inhibitor must have the two following criteria. Firstly, the ILs must be hydrophilic. This would enable them to have access to water molecules. Secondly, functional group introduced to the base cation must be able to create intermolecular hydrogen bonding with water molecules. Based on these two hypotheses they have selected N-(2-hydroxyethyl)-N-methylpyrrolidinium tetrafluoroborate ([HEMP]-BF4 ) and N-butyl-N-methylpyrrolidinium B. Partoon et al. / Fluid Phase Equilibria 337 (2013) 26–31 27 Table 1 Chemical structures and purity of ionic liquids studied in this work. Symbol Chemical name [EMIM]-Cl 1-Ethyl-3-methyl imidazolium chloride Chemical structure Purity/supplier H3C CH3 + N N Cl- 99.8 wt% Sigma–Aldrich OH H3C [OH-C2 MIM]-Cl 2. Materials and methods 2.1. Material [EMIM]-Cl and [OH-C2 MIM]-Cl used in this work are purchased from Sigma–Aldrich. The chemical structures and purity of these ionic liquids are provided in Table 1. Methane gas with a purity of 99.95 mol% is supplied by Merck. Distilled water is used to dilute the ionic liquids to the desired concentrations. 2.2. Apparatus and procedure 2.2.1. Apparatus Hydreval, a motor driven PVT cell, is used in this work [10]. The schematic diagram of the experimental apparatus is shown in Fig. 1. The sapphire chamber is closed at one end by a piston and at the other end by a titanium alloy cell head. The maximum cell capacity is 80 cm3 . Maximum operating pressure of Hydreval is 20 MPa and temperature ranges from 253 K to 523 K. This equipment is equipped with magnetic driven stirrer. Temperature, pressure and volume of reactor are measured and recorded every 2 s with accuracy of ±0.1 K, ±0.01 MPa and ±0.001 cm3 , respectively. These parameters are regulated using Hydreval software. In addition, a camera is attached to monitor and record liquid/gas interface inside the cell. An external pump is used to inject liquid and gas into the sapphire cell. N N 1-Hydroxylethyl-3-methyl-imidazolium chloride tetrafluoroborate ([BMP]-BF4 ) for synthesis and investigation on their inhibition effects. Based on their results, these ILs exhibited good thermodynamic inhibitions at 10 wt% concentration. The results also stated that an optimum combination of anion and cation could lead to good LDHIs. Recently, Li et al. [8] studied the equilibrium conditions of some other ionic liquids. As the literature data on the dual functionality of ionic liquids as hydrate inhibitors is still limited, more investigations are required. In this work, the thermodynamics of methane hydrate formation in the presence of [EMIM]-Cl and [OH-C2 MIM]-Cl are studied. These two ionic liquids are selected for further investigation, as both were reported to have good thermodynamic inhibition effects in the previous studies by Xiao et al. [6] and Li et al. [8] at 10 wt% concentration. The present work is an extension of our previous work [9] on methane hydrate formation at 0.1 wt% and 0.5 wt% concentration of [EMIM]-Cl solution. Our objective is to study the thermodynamic inhibition effect of these ionic liquids on methane gas hydrate at 0.1 wt%, 0.5 wt% and 1 wt% concentrations. In addition, the hydrate equilibrium line for methane in the presence of [EMIM]-Cl is modeled using optimized parameters. + Cl- 99.8 wt% Sigma–Aldrich 2.2.2. Hydrate equilibrium point measurement The equilibrium hydrate formations are measured by isochoric method [1]. The sapphire cell is washed using distilled water and vacuumed. Subsequently, the cell is flushed with methane gas to ensure it is air free. About 40 cm3 of aqueous solution at desired concentration is injected into sapphire cell using the external pump. Then, the cell is cooled down to about 2–3 K above hydrate equilibrium temperature at the desired pressure. The gas is then supplied into the cell to a desired pressure. After the temperature and pressure of the system remains constant, the stirrer is switched on at 600 rpm and the temperature is lowered at a rate of 0.01 K/min to form hydrate. Once hydrate is formed, the temperature of the system is increased with a step method as described by Tohidi et al. [11]. The hydrate formation in the vessel is detected by the T-cycle method [1] as well as visual observation for verification. 3. Theory There are few thermodynamic models for prediction of gas hydrate equilibrium conditions in the presence of electrolyte solutions [12–14]. Most of these models are based on the effect of additives on water activity. The Maddox et al. [15] model for nonelectrolyte inhibitors was used in this work to model the ionic liquid effects on methane hydrate formation phase boundary and details of the derivation of this equation has been described by Pieroen [16]. According to this model, effect of electrolyte on the gas hydrate formation temperature can be explained by: ln(aw ) = −H d nH R 1 T − 1 Tw  (1) where aw is the water activity in electrolyte solution, Hd is the enthalpy of hydrate dissociation, nH is the hydration number, T and Tw are the hydrate formation temperature in electrolyte solution and pure water, respectively, and R is the universal gas constant. Hydrate formation temperature for pure water can be predicted by any hydrate prediction method such as John et al. method [17]. Activity of aqueous electrolyte solutions is predicted using Eqs. (2)–(5) as described by Pitzer and Mayorga [18]. ln(aw ) = −vmMw ϕ + − ϕ ϕ − 1 = |z z |f + m f ϕ = −Aϕ ϕ (2)  2v+ v− v I 1/2 1 + bI 1/2 BMX = ˇ(0) + ˇ(1) exp(−aI 1/2 )  ϕ BMX 2 +m  2(v+ v− ) v 3/2  Cϕ (3) (4) (5) 28 B. Partoon et al. / Fluid Phase Equilibria 337 (2013) 26–31 Fig. 1. Illustration of flow schematic of Hydreval. Table 2 Pitzer and Mayorga parameters for selective salts and ionic liquid. Material NaCl KCl CaCl2 [EMIM]-Cl ˇ(0) ˇ(1) −2 7.650 × 10 4.835 × 10−2 3.159 × 10−1 2.892 × 10−2 Aϕ = 4.0 × 10−6 (T − 273.15)2 + 5.0 × 10−4 (T − 273.15) + 0.3769(6) The change in the enthalpy of hydrate dissociation because of adding electrolytes at two different hydrate temperatures, is considered to depend on the amount of electrolyte and formation of hydrate. Thus, Javanmardi et al. [12] assumed that the enthalpy of dissociation per water molecule in hydrate crystals (Hd /nH R) can be predicted by ionic strength (I) and pressure of the system: (7) In this equation, the effect of electrolytes is included by ionic strength while the effect of hydrate formation was considered by Ref. no. −3 2.664 × 10 2.122 × 10−1 1.614 × 100 1.5648 × 10−1 In these equations, ϕ is the osmotic coefficient, Mw is water molecular weight, v+ and v− are number of ions in the salt formula and z+ and z− are number of cation and anion charges, respectively. Also + + v− , m is the conventional molality and I is the ionic strength v = v (0.5 mi zi ). Pitzer and Mayorga [18] recommended a = 2 and b = 1.2 for all electrolytes. ˇ(0) , ˇ(1) and Cϕ are model parameters. These parameters for selective salts are shown in Table 2. The parameter Aϕ is the Debye–Hückel coefficient. Javanmardi et al. [12] used a value of 0.392 for water at 25 ◦ C in their study; however, Aϕ is a week function of temperature. Thus, in this work, Eq. (6) is used for calculation of Aϕ . This equation is fitted to Debye–Hückel parameter of water reported by Zemaitis et al. [19]. H d q1 I q2 = nH R 1 + q3 P + q4 ln(P) Cf −1 1.270 × 10 −1.800 × 10−3 −3.400 × 10−4 1.013 × 10−2 [18] [18] [18] [21] including pressure term. They optimized parameters of Eq. (7) for electrolytes based on hydrate dissociation condition of some pure gases. These parameters are later optimized by Nasrifar et al. [20] and the same concept is used by them to predict Hd /nH R in the presence of alcohols. In this work, following Javanmardi et al. [12] and Nasrifar et al. [20], these parameters are optimized in the presence of ionic liquids. Optimization is based on the minimization of average absolute error (AAE) for the model predictions, Eq. (8), and experimental hydrate data in the presence of 1 wt% [EMIM]-C1. The optimized parameters are presented in Table 3. N AAE = 1 |Texp . − TCal. |i N (8) i=1 Table 3 Parameters of Eq. (7), optimized based on 1 wt% [EMIM]-Cl hydrate equilibrium data obtained in this work. Electrolytes q1 q2 q3 q4 Ionic liquid Javanmardi et al. [16] Nasrifar et al. [20] This work 597.33 −4.090 × 10−2 2.270 × 10−5 −7.510 × 10−2 1000.0 1.237 × 10−2 −1.205 × 10−2 4.073 × 10−2 222.24 −7.796 × 10−2 3.854 × 10−5 2.530 × 10−2 B. Partoon et al. / Fluid Phase Equilibria 337 (2013) 26–31 Fig. 2. Phase boundary for methane + water in the presence of [EMIM]-Cl at various concentrations. () 1 wt%, () 0.5 wt% [9], () 0.1 wt% [9], (•) pure water [9]. Pitzer and Mayorga [18] parameters for water activity in the presence of [EMIM]-Cl is reported by Zafarani-Moattar and Sarmad [21] and is also provided in Table 2. For calculation of hydrate formation temperature in pure water, John et al. [17] method is used. Unfortunately, Pitzer and Mayorga parameters for [OH-C2 MIM]-Cl are not reported in the literature and thus this model cannot be employed for this ionic liquid. 4. Results and discussion 4.1. Phase behavior in the hydrate forming region In our previous work, we showed that the measured hydrate formation conditions using Hydreval are in good agreement with the literature data [9]. The phase equilibrium data for [EMIM]Cl and [OH-C2 MIM]-Cl at different concentrations are shown in Figs. 2 and 3, respectively. In the literature, the reported equilibrium data for these ionic liquids are at high concentration of 10 wt% whereas in this work the concentration of these ionic liquids is in the range of 0.1–1 wt% [6,8]. The measured equilibrium data for these systems are provided in Table 4. As expected from the behavior of thermodynamic inhibitors the equilibrium line of methane hydrate is shifted to the lower temperatures in the presence of ionic liquids as shown in Figs. 2 and 3. Moreover, the inhibition effect of these ionic liquids on the equilibrium line is small at low pressures and it becomes more significant Fig. 3. Phase boundary for methane + water in the presence of [OH-C2 MIM]-Cl at various concentrations. () 1 wt%, () 0.5 wt%, () 0.1 wt%, (•) pure water [9]. 29 Fig. 4. Comparison of [EMIM]-Cl and [OH-C2 MIM]-Cl effects with normal electrolyte on methane hydrate phase boundaries. () 1 wt% [OH-C2 MIM]-Cl, () 1.0 wt% [EMIM]-Cl, () 5.0 wt% NaCl [22], (×) 5.0 wt% KCl [22], (+) 5.0 wt% CaCl2 [22], (䊉) pure water [9]. at higher pressures. The same behavior is exhibited by the data published by Xiao et al. [6]. In addition, thermodynamic inhibition effect is increased when the concentration of ionic liquid in the aqueous solution increases. Fig. 4 compares the influence of [EMIM]-Cl and [OH-C2 MIM]-Cl at 1 wt% on methane hydrate equilibrium. [OH-C2 MIM]-Cl shows better inhibition effects than [EMIM]-Cl. The same behavior is reported by Li et al. [8]. This may be attributed to the presence of hydroxyl group in [OH-C2 MIM]-Cl which can form hydrogen bonding with water molecules and thus it shows better hydrate inhibiting effect rather than [EMIM]-Cl. However, this effect is not too significant. In addition to the measured data, the data for methane hydrate equilibrium line in the presence of 5 wt % salts in aqueous solutions are presented in Fig. 4. These data are obtained from Mohammadi et al. [22] and included for comparison. As shown in Fig. 4, the thermodynamic inhibition effect of these ionic liquids is slightly lower than that of the salts. This may due to lower concentration of the ionic liquids or higher activity coefficients of these ionic liquids as shown in Fig. 5. The mean ionic activity coefficient ( ±) of electrolyte presented in Fig. 5 is calculated using Pitzer and Mayorga method [18]. The higher activity coefficient of [EMIM]Cl at various concentrations shows less non-ideality behavior and therefore, its inhibition effect should be less than normal electrolytes. The activity coefficient for [OH-C2 MIM]-Cl is not calculated due to lack of the model parameter of this ionic liquid. Fig. 5. Comparison of [EMIM]-Cl, KCl, NaCl and CaCl2 activity as a function of concentration. () NaCl, () KCl, (•)CaCl2 , ()[EMIM]-Cl. 30 B. Partoon et al. / Fluid Phase Equilibria 337 (2013) 26–31 Table 4 Methane hydrates equilibrium data in the presence of ionic liquids, obtained in this work. Ionic liquid Concentration (wt%) 0.1 [EMIM]-Cl 0.5 1.0 0.1 [OH-C2 MIM]-Cl 0.5 1.0 a b Pa (MPa) Tb (K) P (MPa) T (K) P (MPa) T (K) P (MPa) T (K) 3.88 4.21 4.43 3.89 4.01 4.43 4.20 4.55 4.95 276.8 277.2 278.0 276.9 277.3 278.4 277.1 278.2 279.3 4.79 5.33 5.64 4.81 5.33 5.84 5.68 6.30 7.21 279.3 280.2 281.2 279.3 280.2 281.1 280.3 281.2 282.3 6.47 7.90 8.66 6.89 8.01 8.70 8.10 9.13 10.17 282.2 283.2 284.2 282.2 283.3 284.2 283.1 284.2 285.3 9.25 10.48 11.29 9.49 10.53 11.43 11.04 12.16 285.2 286.3 287.3 285.2 286.2 287.3 286.2 287.3 4.18 4.32 4.88 4.28 4.31 4.88 4.49 4.80 4.83 276.7 277.3 278.5 276.8 277.2 278.0 276.7 277.3 278.5 5.00 5.77 6.29 4.99 5.77 6.29 5.21 6.16 6.84 278.9 280.4 281.2 279.3 280.2 281.4 278.9 280.2 281.2 6.87 7.84 9.00 6.87 8.04 9.00 7.34 7.91 9.21 282.3 283.2 284.4 282.2 283.3 284.2 282.3 282.9 284.4 9.24 10.39 11.26 9.38 10.58 11.60 9.93 11.55 12.33 285.1 286.2 287.1 285.2 286.3 287.3 285.1 286.2 287.1 Pressure uncertainty is ±0.01 MPa. Temperature uncertainty is ±0.1 K. Table 5 Prediction results of methane hydrate equilibrium line in the presence of salts and [EMIM]-Cl. NaCl KCl CaCl2 [EMIM]-Cl [EMIM]-Cl [EMIM]-Cl Concentration (wt%) No. data point 5 5 5 0.1 0.5 1 5 6 6 12 12 11 Temperature range 274.2–283.6 271.6–283.2 272.0–283.0 276.8–287.3 276.8–287.3 277.1–287.3 AAE (K) Ref. 0.1 0.5 0.6 0.6 0.3 0.2 [22] [22] [22] [9] [9] This work 4.2. Modeling Hydrate equilibrium conditions in the presence of selective salts are modeled and compared to the literature data [22] to assess the applicability of this model. The results are tabulated in Table 5. As shown in Table 5, the predicted results are in agreement with reported data with AAE less than 0.6. Therefore, the model is used for prediction of hydrate equilibrium condition in the presence of [EMIM]-Cl. The dependency of hydrate dissociation enthalpy to the ionic strength is calculated by minimizing the AAE, Eq. (8), of methane hydrate equilibrium temperature at 1 wt% of [EMIM]-Cl. Then the optimized parameters are used to predict hydrate equilibrium temperatures of methane hydrate in 0.5 wt% and 0.1 wt% of [EMIM]-Cl solutions. Results of these calculations are also presented in Table 5 and illustrated in Fig. 6. As shown in Fig. 6, the model prediction is in good agreement with experimental data. By taking into consideration that the model parameters are only optimized for methane hydrate in 1 wt% of [EMIM]-Cl solution, the model is capable to predict hydrate equilibrium condition at other [EMIM]-Cl concentrations with AAE less than 0.6. In addition, prediction results and experimental measurements show the same trend of the equilibrium behavior with regard to pressure changes. At low pressure, the equilibrium line of methane hydrates in the presence of ionic liquids slightly shifted to lower temperature compare to methane + pure water hydrate equilibrium line, while at higher pressure this transformation becomes more significant. 5. Conclusion The equilibrium conditions of methane hydrate in the presence of [EMIM]-Cl and [OH-C2 MIM]-Cl are measured in this work. The thermodynamic inhibition effects on methane hydrate equilibrium in the presence of these ILs are less significant at low pressures. However, when the pressure is higher than 7 MPa, their thermodynamic inhibition effects are more significant. Furthermore, a thermodynamic model is optimized for prediction of gas hydrate equilibrium temperature in the presence of ionic liquids and the prediction results are in good agreement with data produced in this work. List of symbols Fig. 6. Experimental and predicted methane hydrate formation temperature for various [EMIM]-Cl solutions. () 1 wt% [EMIM]-Cl, () 0.5 wt% [EMIM]-Cl [9], () 0.1 wt% [EMIM]-Cl [9], (•) pure water [9], ( · · )1 wt% [EMIM]-Cl prediction, ( · ) ) pure water 0.5 wt% [EMIM]-Cl prediction, (· · ·) 0.1 wt% [EMIM]-Cl prediction, ( prediction. AAE Aϕ Cϕ H I Mw N P R T average absolute error Debye–Hückel parameter parameter of Pitzer and Mayorga model enthalpy, J ionic strength, mol molecular weight of water total number of data points pressure, kPa universal gas constant, 8.314 J/(mol K) temperature, K B. Partoon et al. / Fluid Phase Equilibria 337 (2013) 26–31 V volume, m3 Z compressibility factor aw water activity A parameter of Eq. (7) fixed as 2 B parameter of Eq. (6) fixed as 1.2 M molality, mol N number of moles, mol nH hydration number q1 , q2 , q3 , q4 parameters of Eq. (7) V number of ions in salt or ionic liquid formula Z charge of ions in salt or ionic liquid formula Greek letters ˇ(0) , ˇ(1) parameters of Pitzer and Mayorga model ϕ osmotic coefficient ± mean ionic activity coefficient Subscripts and superscripts + cation − anion Cal. calculated dissociation d exp. experimental f moles of gas at any time i initial moles of gas MX mixed water w ϕ osmotic coefficient related parameter Acknowledgment This study was supported by Ministry of Science, Technology and Innovation of Malaysia under e-science project Grant No. 0602-02-SF0076. 31 References [1] E.D. Sloan, C. Koh, Clathrate Hydrates of Natural Gases, 3rd ed., CRC Press-Taylor & Francis Group, New York, 2008. [2] E.G. Hammerschmidt, Ind. Eng. Chem. 26 (1934) 851–855. [3] J.J. Carroll, Natural Gas Hydrates: A Guide for Engineers, 2nd ed., Gulf Professional Pub., 2009. [4] C.W. Xiao, H. Adidharma, Chem. Eng. Sci. 64 (2009) 1522–1527. [5] L. Del Villano, M.A. Kelland, Chem. Eng. Sci. 65 (2010) 5366–5372. [6] C.W. Xiao, N. Wibisono, H. Adidharma, Chem. Eng. Sci. 65 (2010) 3080–3087. [7] K.S. Kim, J.W. Kang, S.P. Kang, Chem. Commun. 47 (2011) 6341–6343. [8] X.S. Li, Y.J. Liu, Z.Y. Zeng, Z.Y. Chen, G. Li, H.J. Wu, J. Chem. Eng. Data 56 (2011) 119–123. [9] B. Partoon, N.M.S. Wong, S.N.A. Malik, K.M. Sabil, Experimental study of methane hydrate equilibrium and formation kinetic in the presence of an ionic liquid, in: International Conference on Process Engineering and Advanced Materials, Kuala Lumpur, Malaysia, 2012. [10] Anonymous, Hydrates Studies System (HYDREVAL), http://www.vincitechnologies.com/products-explo.aspx?IDP=82288&idr=113221&IDR2=88545 &IDM=601177 (accessed 25.06.12). [11] B. Tohidi, R.W. Burgass, A. Danesh, K.K. Østergaard, A.C. Todd, Improving the accuracy of gas hydrate dissociation point measurements, in: Gas Hydrates: Challenges for the Future, New York, 2000. [12] J. Javanmardi, M. Moshfeghian, R.N. Maddox, Energy Fuels 12 (1998) 219–222. [13] J. Javanmardi, M. Moshfeghian, R.N. Maddox, Gas Hydrates: Challenges for the Future, vol. 912, 2000, pp. 713–721. [14] H. Najibi, A. Chapoy, H. Haghighi, B. Tohidi, Fluid Phase Equilib. 275 (2009) 127–131. [15] R.N. Maddox, M. Moshfeghian, E. Lopez, C.H. Tu, A. Shariat, J. Flynn, Predicting hydrate temperature at high inhibitor concentration, in: Laurance Reid Gas Conditioning Conference, Norman, OK, 1991, pp. 273–294. [16] A.P. Pieroen, Recl. Trav. Chim. Pays-Bas 74 (1955) 995–1002. [17] V.T. John, K.D. Papadopoulos, G.D. Holder, AlChE J. 31 (1985) 252–259. [18] K.H. Pitzer, G. Mayorga, J. Phys. Chem. 77 (1973) 2300–2307. [19] J.E. Zemaitis, D.M. Clark, M. Rafal, N.C. Scrivner, Handbook of Aqueous Electrolyte Thermodynamics: Theory and Application, John Wiley & Sons, New York, 1986. [20] K. Nasrifar, M. Moshfeghian, R.N. Maddox, Fluid Phase Equilib. 146 (1998) 1–13. [21] M.T. Zafarani-Moattar, S. Sarmad, CALPHAD 35 (2011) 331–341. [22] A.H. Mohammadi, W. Afzal, D. Richon, J. Chem. Thermodyn. 40 (2008) 1693–1697.